FreeStyle Libre kontinuerliga glukosmätare (CGM) kan maniska risken för patienter med T1DM att minska att bli inlagda på sjukhus för kardiovaskulära komplikationer,
enligt studie från NDR, publicerad i Diabetologia
med indata från 15 000 patienter.
Data utgick från från det svenska nationella diabetesregistret och det svenska nationella patientregistret.
Av 14 829 vuxna med typ 1-diabetes hade cirka 0,8 % tidigare haft en allvarlig hypoglykemi (SHE) vid studiens start.
Efter två år skedde en 78-procentig minskning av hjärtsjukdomsrelaterade sjukhusinläggningar för patienter med en historia av SHE som använde FreeStyle Libre-tekniken
i motsats till de som använde traditionell blodsockerövervakning (BGM).
Studiens författare noterade att fördelar också sågs hos patienter som använde en FreeStyle Libre CGM utan tidigare historia av SHE. Om man till exempel tittar på patienter utan tidigare historia av hjärt-kärlsjukdom, var användning av CGM-tekniken fortfarande kopplad till 49 % minskning av sjukhusinläggningar på grund av hjärtkomplikationer jämfört med BGM.
”CGM ger människor möjlighet att proaktivt hantera sin diabetes och göra välgrundade hälsoval genom ständig feedback i realtid om sina glukosnivåer”, säger medförfattaren David Nathanson, MD, PhD, forskare vid Karolinska Universitetssjukhuset i Sverige, i ett uttalande. ”Dessa data visar att användning av CGM är kopplat till betydligt färre sjukhusinläggningar relaterade till hjärtproblem, vilket kan ha en betydande inverkan på patienter, deras familjer och sjukvårdssystemet genom att lindra medicinska, känslomässiga och ekonomiska utmaningar, och med hjärtfördelar gör Libre det lättare för människor att ta kontroll över sin hälsa
Läs hela studien här.
https://link.springer.com/article/10.1007/s00125-025-06438-y
Severe hypoglycaemia is associated with increased risk of adverse cardiovascular complications in adults with type 1 diabetes: risk mitigation using intermittently scanned continuous glucose monitoring
- Katarina Eeg-Olofsson,
- David Nathanson,
- Tim Spelman,
- Mattias Kyhlstedt,
- Alexander Seibold,
- Fleur Levrat-Guillen &
- Jan Bolinder
Abstract
Aims/hypothesis
It has been proposed that severe hypoglycaemia events (SHE) may increase the risk of adverse CVD complications in adults with type 1 diabetes.
The aim of this study was to evaluate the risk of CVD complications following SHE in a large cohort of adults with type 1 diabetes, and to compare the risk of post-SHE CVD complications for users of intermittently scanned continuous glucose monitoring (isCGM) vs users of blood glucose monitoring (BGM).
Methods
This comparative retrospective cohort study used data from the Swedish National Diabetes Register and the Swedish National Patient Register.
We identified people with type 1 diabetes who had a hospitalisation for CVD complications.
Rates of hospitalisation were compared between those with an index SHE and those without, and within isCGM or BGM subgroups.
The study baseline was date of the first SHE prior to the isCGM index date.
Results
We identified 14,829 adults with type 1 diabetes with up to 2 years of follow-up, of which 1313 had an index SHE.
In the full cohort, the relative rate of hospitalisations for CVD complications was 2.06-fold (95% CI 1.48, 2.85) in those with prior SHE.
Of these 1313 participants with prior SHE, 970 were using isCGM and 343 were using BGM.
Hospitalisations for post-SHE CVD complications were significantly lower for isCGM users (5.40 per 100 person-years of follow-up; 95% CI 4.59, 6.31) compared with BGM control participants (14.23 per 100 person-years of follow-up; 95% CI 11.95, 16.82), which represents a 78% relative reduction in rates of post-SHE CVD complications for isCGM users (relative rate 0.22; 95% CI 0.11, 0.43; p<0.001), after adjustment for confounders.
Conclusions/interpretation
In adults with type 1 diabetes, SHE is associated with an increased risk of hospitalisation for adverse CVD complications.
This risk is significantly reduced in isCGM users compared with BGM control participants.
From the article
Introduction
Despite advances in glycaemic management, people with type 1 diabetes remain at higher risk of CVD and CVD-related mortality than people without diabetes [1]. Data from the Swedish National Diabetes Register (NDR) reveal that people with type 1 diabetes and HbA1c ≤52 mmol/mol (≤6.9%) have a twofold greater risk of CVD death than a matched population without diabetes [2]. For individuals with type 1 diabetes and HbA1c ≥82 mmol/mol (≥9.7%), this risk is estimated to be 8–10-fold [2].
The Diabetes Control and Complications Trial (DCCT) and the Epidemiology of Diabetes Interventions and Complications (EDIC) study have shown that more-intensive glucose control for people with type 1 diabetes is associated with a 42% reduced risk of any cardiovascular event and a 57% risk reduction of death from CVD [3] over a mean of 6.5 years. However, prior severe hypoglycaemia events (SHEs) are reported to increase the risk of adverse CVD complications in people with type 1 diabetes [4], and SHEs are independently associated with accelerated atherosclerosis, increased risk of CVD events and all-cause mortality [5,6,7]. Acute or recurrent hypoglycaemia episodes are known to induce proinflammatory mediators, platelet activation and markers of vascular endothelial stress in type 1 diabetes [8, 9].
Continuous glucose monitoring (CGM) technologies have enabled people with type 1 diabetes to significantly improve their glucose levels and reduce exposure to hypoglycaemia, compared with using blood glucose monitoring (BGM) test strips [10,11,12,13]. We have previously shown that the use of intermittently scanned CGM (isCGM) is associated with lowered HbA1c within 6 months of initiation, and that these levels persist for at least 2 years of follow-up, in addition to lower rates of SHEs compared with BGM control participants [14]. Using data from the NDR, linked with hospital admissions data from the Swedish National Patient Register (NPR), we have also reported that, compared with a matched cohort of BGM users, isCGM users with type 1 diabetes have a 36–75% relative reduction in rates of hospitalisation for CVD events, including acute myocardial infarction (AMI), stroke, heart failure and atrial fibrillation, as well as a relative reduction in hospitalisation for acute diabetes events (68% relative reduction in rates of SHE and 45% relative reduction for diabetic ketoacidosis) over a 24-month follow-up period, together with an HbA1c reduction of approximately 0.3% [15].
Mechanistically, it seems unlikely that reductions in CVD events were linked to improved glucose control, considering the short observation period and modest difference in HbA1c. Instead, we speculate that the reductions in CVD events may be associated with the observed risk reduction for SHEs. Studies in healthy individuals and in people with type 1 diabetes indicate that acute hypoglycaemia leads to increased proinflammatory cytokines, mobilisation of inflammatory leukocytes and increased platelet reactivity [16,17,18]. Such consequences are hypothesised to promote atherogenesis and plaque instability in people with diabetes. To investigate this hypothesis, the current study uses linked data from the NDR and NPR to examine the association between prior SHE and hospital admissions for non-fatal CVD complications and CVD-related death in adults with type 1 diabetes, and compared the rate of CVD complications following a prior SHE for isCGM users vs a control group of BGM users.
Discussion
This comparative retrospective study demonstrates that prior SHE is associated with an increased risk of hospitalisation for CVD complications in adults with type 1 diabetes, and reveals important changes in the risk of post-SHE CVD complications for isCGM users. Our data confirm that risk of hospital admission for CVD complications is approximately doubled among adults with type 1 diabetes who had a previous SHE, compared to those without a recorded SHE (Fig. 1a). The highest rate of hospital admissions for CVD complications was in the cohort of adults with type 1 diabetes who were BGM users (Fig. 1b). Adults with type 1 diabetes who were isCGM users had a lower rate of hospitalisation for CVD complications (Fig. 1c), in both the group with and those without a prior SHE. In keeping with the wider observation that a prior SHE episode confers an approximately doubled risk of hospitalisation for CVD complications, the post-SHE cohort of isCGM users exhibited a 1.89-fold relative increase in rate of admission for CVD complications (Fig. 1c).
The most notable outcome is that isCGM users with a prior SHE had a greatly reduced frequency of admission for CVD complications compared to BGM users with a previous SHE (Fig. 2). Equally notable are the significant reductions in risk of CVD complications for isCGM users with no prior SHE, compared with BGM users. Use of isCGM has been shown to reduce less-severe hypoglycaemia in type 1 diabetes [12, 13], which may contribute to the lower incidence of CVD complications shown here. These outcomes were replicated when using a narrower composite definition of CVD complications, including AMI, stroke and cardiovascular death (ESM Fig. 2). This aligns with our previous study showing that isCGM users have a significant relative reduction in rates of hospital admissions for cardiovascular complications of type 1 diabetes, compared with BGM control participants, including stroke, AMI, heart failure and atrial fibrillation [15]. Our subgroup analysis showed that risk of CVD complications was significantly higher in the subgroup of adults with type 1 diabetes and a history of CVD (Fig. 3), and that using isCGM significantly reduced this risk.
Our study has several noteworthy outcomes.
These include demonstration of the association of at least one prior SHE with a significantly increased risk of CVD complications for people with type 1 diabetes, using the composite endpoint of hospitalisation for CVD complications, rather than CVD mortality alone. A 2014 study using NDR data to show an association of prior SHE with CVD mortality in type 1 diabetes [4] concluded that, although prior SHE was associated with reduced survival following a major CVD event, it was not associated with an increased risk of a subsequent fatal CVD event. Our data, using hospital admissions data, suggest an association between prior SHE and an increased risk of composite CVD complications, for people with type 1 diabetes using either BGM or isCGM. This finding must be reconciled with the EURODIAB study outcomes, which did not find that prior SHE increased the risk of fatal or non-fatal cardiovascular events [24] and the finding of a prospective observational study on two cohorts of people with type 1 diabetes, one Danish (n=269) and one Dutch (n=482), neither of which found an association between episodes of SHE or impaired awareness of hypoglycaemia and CVD mortality when not part of a composite outcome [25]. The difference in outcomes between the previous NDR study on CVD mortality following SHE [4] and the current study may also reflect different selection criteria and reporting aims. The 2014 study included risk factor data from the NDR between January 1987 and December 2010, and reported risk equations for survival after myocardial infarction and stroke only, in 1839 adults with type 1 diabetes. Thus, their conclusion that prior SHE was not associated with an increased risk of subsequent CVD was based only on myocardial infarction and stroke. Our study used a composite outcome comprising occurrence of hospital admissions with a primary diagnosis of non-fatal AMI, coronary heart disease, non-fatal stroke, atrial fibrillation or heart failure, or recorded cardiovascular death. The different methodology and the wider selection of CVD-related primary diagnoses is likely to influence comparisons with earlier studies, as each of these primary diagnoses have been associated with increased hospitalisation for adults with type 1 diabetes using BGM, and with reductions in relative rates for isCGM users [15]. An assessment of atherosclerosis in type 1 diabetes in the DCCT cohort found that SHE is associated with increased coronary artery calcification scores for individuals with HbA1c <58 mmol/mol (<7.5%) [26]. During acute hypoglycaemia, heart rate, systolic BP and blood flow (which are associated with risk of CVD) all increase [27].
The association with atherosclerotic disease is also supported by one study showing that repeated hypoglycaemia in type 1 diabetes, including in individuals with impaired awareness of hypoglycaemia, may be an aggravating factor for markers of preclinical atherosclerosis [6]. Prior SHE may also be a marker for increased frequency of non-severe hypoglycaemia, which has a cumulative effect on cardiac structures, as hypoglycaemia is implicated in cardiac arrythmias in type 1 diabetes [28,29,30]. More directly, studies have shown that acute hypoglycaemia can have proinflammatory sequelae that may contribute to atherogenesis in type 1 diabetes [16, 17]. Repeated hypoglycaemia is also associated with enhanced coagulation, oxidative stress, vascular inflammation, endothelial dysfunction and platelet activation [31], which are features of CVD. However, we should point out that isCGM has also been shown to reduce time in hyperglycaemia and glucose variability [12, 13], which are not recorded in the NDR. The influence of these variables on CVD complications should be examined in future trials.
The association of prior SHE with increased CVD outcomes in type 1 diabetes is a clinically important finding of the current and previous studies [4, 15] using the NDR and NPR datasets. Notably, our findings do not prove a causal link between severe or recurrent hypoglycaemia and CVD complications but only an association between the two phenomena. However, in our study, the use of isCGM was certainly associated with a lower risk of CVD complications as a composite outcome in adults with type 1 diabetes, both for those with and without a prior SHE, compared with use of BGM, with the reduction being most noticeable in those with a prior SHE (Fig. 2).
Strengths and limitations
This retrospective cohort study has limitations. Initiation of isCGM is a key variable in our analysis, but other factors may have influenced the outcomes.
For example, it is possible that new isCGM users received device-related education at the point of initiation, which may have improved diabetes or cardiovascular self-care behaviours that we cannot control for. Equally, the NDR does not contain information on glycaemic variability or time in hyperglycaemia, so their influence cannot be assessed here. When comparing baseline characteristics of isCGM users against BGM control participants before weighting (ESM Table 2), there was an age differential between isCGM users and BGM control participants (mean 50.08 vs 66.21 years) and a minority of adults with type 1 diabetes were CSII users (12.6% of isCGM users and 2.3% of BGM control participants). Both age and CSII therapy may be associated with reduced CVD mortality compared with multiple daily injections with insulin [32], and may represent a confounding factor within the composite CVD complications reported. Likewise, the baseline prevalence of ischaemic heart disease was lower among isCGM users (8.6%) than BGM control participants (15.7%).
However, after PS-IPTW, the weighted standard differences between the groups with regard to age (−0.172), use of CSII (0.167) and ischaemic heart disease (−0.135) were all below 0.20 (ESM Table 2), which suggests adequate balancing of the two groups in each case, with only a small degree of residual confounding after weighting that is unlikely to affect the outcomes in a meaningful way [22]. The same may be argued for the baseline prevalence of kidney disease, where the weighted standardised difference for serum creatinine between the cohorts was above 0.20. However, a double robust method, using additional variable adjustment in the regression analysis, was used to minimise the effect of this remaining imbalance, and other measures of kidney function were adequately balanced. Thus, we believe that kidney disease is unlikely to influence the outcomes meaningfully. The majority of other covariates between isCGM and BGM users were associated with weighted standard differences below 0.10, with negligible confounding influence (ESM Table 2). PS-IPTW also indicates that the isCGM and BGM cohorts were well matched for baseline covariates whether they had a prior SHE or not (Table 1, ESM Tables 3 and 4). These analytical validations increase the internal confidence of this comparative study, although we acknowledge that this is a limitation for the generalisability of our outcomes. Although the PS-IPTW assessment gives us confidence that the BGM and isCGM cohorts are adequately balanced, the NDR provides no information as to why a person with type 1 diabetes does not use either isCGM or real-time CGM, which is a limitation for interpretation of the comparative outcomes. As we have used PS-IPTW to balance baseline characteristics for BP and lipid profiles, along with physical activity measures and any overt ischaemic heart disease, no information on CVD co-medications is included. As the NDR does not report specific dates related to SHEs, we cannot delineate the exact time periods between occurrence of SHEs and CVD complications, which is a limitation.
Baseline characteristics related to ethnicity, education or indices of social deprivation are not available in the NDR or NPR, so the influence of these factors on admission rates for CVD complications cannot be assessed. These limitations allow us to make clear associations between the use of isCGM with regard to CVD complications in adults with type 1 diabetes but do not allow us to draw definitive conclusions regarding causality. Strengths of our study include the large unselected population, with almost complete coverage of the adult type 1 diabetes population in Sweden, together with the availability of a matched control group of type 1 diabetes patients with previous SHE who used BGM rather than isCGM or real-time CGM systems. This has allowed us to minimise the risk of residual confounding and to mitigate the impact of other factors on glycaemic control over the study period that are independent of isCGM use.
Conclusions
This retrospective cohort study shows that a prior SHE in adults with type 1 diabetes in Sweden is associated with an approximate doubling of risk of hospitalisation for CVD complications, compared to adults with type 1 diabetes and no prior SHE. Use of isCGM by adults with type 1 diabetes is associated with a 78% relative reduction in the rate of hospitalisation for CVD complications following a SHE, compared to adults with type 1 diabetes who use BGM.
These outcomes illustrate the impact of SHE on the incidence of CVD episodes requiring hospitalisation, which may impact the long-term cost-effectiveness of using isCGM for the management of glucose levels in type 1 diabetes.
Läs hela studien pdf free
https://link.springer.com/article/10.1007/s00125-025-06438-y
Nyhetsinfo
www red DiabetologNytt